PEDIGREE OF CCELENTERA 207 



conception, with which Lyell first impressed the world, of the uniformity 

 of natural processes throughout the long history of the earth. Thus in 

 connection with Coelentera we learn that there were great coral reefs in 

 the incalculably distant past, just as there are coral reefs still. So in 

 the Cambrian rocks, which are next to the oldest, there are on sandy 

 slabs markings exactly like those which are now left for a few hours 

 when a large jelly-fish stranded on the flat beach slowly melts away. 

 On the other hand, some forms of life which lived long ago seem to 

 have been very different from any that now remain, as is well shown 

 by the abundant Graptolite fossils, which, though probably Coelentera, 

 do not fit well into any of the modern classes. 



As to the pedigree of the Coelentera, the facts of individual life- 

 history, and the scientific imagination of naturalists, help us to construct 

 a genealogical tree — a hypothetical statement of the case. Thus it 



Fig. 112. — Hydroctena. A medusoid with suggestions 

 of Ctenophore structure, but a true medusoid none the less. 



ab.o., Aboral sensory organ ; T., retractile tentacle ; 

 v., velum ; M., month ; ST., stomach. 



seems very Ukely that the ancestral many-celled animals — ancestral to 

 Sponges, Coelentera, and all the rest — were small two-layered tubular 

 or oval forms. The many-celled animals must have begun as clusters 

 of cells ; the question is, what sort of clusters — spheres of one layer of 

 cells, or mouthless ovals, or little discs of cells, or two-layered thimble- 

 like sacs ? Possibly there were many forms, but Haeckel and other 

 naturalists were led to fix their attention especially on the two-layered 

 sac or gastrula, because this form keeps continually cropping up as an 

 embryonic stage in the life-history of animals, whether sponge or coral, 

 earthworm or starfish, mollusc or even vertebrate, and also because this 

 is virtually the form which is exhibited by the simx^lest sponges 

 (Ascons), the simplest Coelentera {Hydra), and even by the simplest 

 " worms " (Turbellarians). 



If we begin in our survey with such a gastrula-like ancestor, the 

 probabilities are certainly in favour of the supposition that it was a free- 



